Deep Neural Network for the Automated Detection and Diagnosis of Seizure using EEG Signals

Abstract

An encephalogram (EEG) is generally used ancillary test for the diagnosis of epilepsy. The EEG signal contains information about brain electrical activity. Neurologists employ direct visual inspection to identify epileptiform abnormalities and this technique can be timeconsuming which provides variable results secondary to reader expertise level and is limited to identify the abnormalities. Since it is essential to develop a computer-aided diagnosis (CAD) system to automatically distinguish the classes of EEG signals using machine learning techniques. This is the first time to study the convolutional neural network (CNN) for analysis of EEG signals. In this work, 11-layer deep convolutional neural network (CNN) algorithm is implemented to detect normal, preictal, and seizure classes. This technique achieved an accuracy as high as possible with 99%